Merge pull request #58 from SylphAI-Inc/li

Li

class_hierarchy_edges.csv

@@ -0,0 +1,68 @@
+Component,ListParser
+Component,JsonParser
+Component,YamlParser
+Component,ToolManager
+Component,Prompt
+Component,ModelClient
+Component,Retriever
+Component,FunctionTool
+Component,Tokenizer
+Component,Generator
+Component,Embedder
+Component,BatchEmbedder
+Component,Sequential
+Component,FunComponent
+Component,ReActAgent
+Component,OutputParser
+Component,TextSplitter
+Component,DocumentSplitter
+Component,ToEmbeddings
+Component,RetrieverOutputToContextStr
+Component,DefaultLLMJudge
+Component,LLMAugmenter
+Generic,LocalDB
+Generic,Retriever
+Generic,GeneratorOutput
+Generic,Parameter
+Generic,Sample
+Generic,Sampler
+Generic,RandomSampler
+Generic,ClassSampler
+ModelClient,CohereAPIClient
+ModelClient,TransformersClient
+ModelClient,GroqAPIClient
+ModelClient,GoogleGenAIClient
+ModelClient,OpenAIClient
+ModelClient,AnthropicAPIClient
+Retriever,BM25Retriever
+Retriever,PostgresRetriever
+Retriever,RerankerRetriever
+Retriever,LLMRetriever
+Retriever,FAISSRetriever
+Enum,DataClassFormatType
+Enum,ModelType
+Enum,DistanceToOperator
+Enum,OptionalPackages
+DataClass,EmbedderOutput
+DataClass,GeneratorOutput
+DataClass,RetrieverOutput
+DataClass,FunctionDefinition
+DataClass,Function
+DataClass,FunctionExpression
+DataClass,FunctionOutput
+DataClass,StepOutput
+DataClass,Document
+DataClass,DialogTurn
+DataClass,Instruction
+DataClass,GeneratorStatesRecord
+DataClass,GeneratorCallRecord
+Generator,CoTGenerator
+Generator,CoTGeneratorWithJsonOutput
+OutputParser,YamlOutputParser
+OutputParser,JsonOutputParser
+OutputParser,ListOutputParser
+OutputParser,BooleanOutputParser
+Optimizer,BootstrapFewShot
+Optimizer,LLMOptimizer
+Sampler,RandomSampler
+Sampler,ClassSampler

docs/source/_static/css/custom.css

@@ -10,6 +10,20 @@
 .theme-version {
   display: none;
 }
+.bd-sidebar-primary {
+  width: 100%; /* Full width by default */
+}
+
+/* Adjust width for larger screens */
+@media (min-width: 768px) { /* Example breakpoint for tablets and larger */
+  .bd-sidebar-primary {
+      width: 270px; /* Adjust the width for larger screens */
+  }
+}
+
+/* .copyright {
+  text-align: center;
+} */
 p {
     font-size: 0.9em;
     margin-bottom: 1.15rem;

docs/source/developer_notes/base_data_class.rst

@@ -12,7 +12,7 @@ Tensor wraps a multi-dimensional matrix to better support its operations and com
 In LLM applications, data constantly needs to interact with LLMs in the form of strings via prompt and be parsed back to structured data from LLMs' text prediction.
 :class:`core.base_data_class.DataClass` is designed to ease the data interaction with LLMs via prompt(input) and text prediction(output).
 
-.. figure:: /_static/images/dataclass.png
+.. figure:: ../_static/images/dataclass.png
     :align: center
     :alt: DataClass
     :width: 680px

docs/source/developer_notes/class_hierarchy.rst

@@ -0,0 +1,6 @@
+LightRAG Class Hierarchy Visualization
+=============================
+
+.. raw:: html
+
+    <iframe src="../_static/class_hierarchy.html" width="100%" height="750px"></iframe>

docs/source/developer_notes/evaluation.rst

@@ -1,4 +1,4 @@
-A Guideline on LLM Evaluation
+LLM Evaluation
 ====================================
 
 .. admonition:: Author

docs/source/developer_notes/optimizer.rst

@@ -1,2 +1,2 @@
-Optimizer - The Future of LLM applications
-==========================================================
+Optimizer
+==========================================================

docs/source/index.rst

@@ -66,23 +66,29 @@ It is light, modular, and robust.
                return await self.generator.acall({"input_str": query})
 
 
-         qa = SimpleQA()
-         answer = qa("What is LightRAG?")
 
 
 
 
 Clarity and Simplicity
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
+Developers who are building real-world Large Language Model (LLM) applications are the real heroes.
+As a library, we provide them the foundamental building blocks with 100% clarity and simplicity.
 
-We understand that developers building real-world Large Language Model (LLM) applications are the real heroes. Just like AI researchers and engineers who build models on top of PyTorch, developers require **Maximum Flexibility and Customizability**: Each developer has unique data needs to build their own models/components, experiment with In-context Learning (ICL) or model finetuning, and deploy the LLM applications to production. This means the library must provide fundamental lower-level building blocks and strive for clarity and simplicity:
+.. We support them with require **Maximum Flexibility and Customizability**:
 
-- We maintain no more than two levels of subclasses.
-- Each core abstract class is designed to be robust and flexible.
-- We use 10X less code than other libraries to achieve 10X more robustness and flexibility.
+.. Each developer has unique data needs to build their own models/components, experiment with In-context Learning (ICL) or model finetuning, and deploy the LLM applications to production. This means the library must provide fundamental lower-level building blocks and strive for clarity and simplicity:
 
-Here is the printed out structure of ``qa``:
+- Two foundamental and powerful base classes: ``component`` for the pipeline and ``DataClass`` for the data interaction with LLMs.
+- We end up with less than two levels of subclasses. :doc:`developer_notes/class_hierarchy`.
+- The result is a library with bare minimum abstraction with maximum flexibility and customizability.
+
+.. - We use 10X less code than other libraries to achieve 10X more robustness and flexibility.
+
+.. - `Class Hierarchy Visualization <developer_notes/class_hierarchy.html>`_
+
+Similar to PyTorch module, our ``component`` gives us a great visualization on the pipeline structure.
 
 .. code-block::